def bwt_command(filename):
"""
Ecrit à l'écran la transformée de Burrows-Wheeler du génome contenu
dans le fichier FASTA.
"""
seq = FastaFile(filename).complete_sequence()
print bwt(seq)
def main():
start = time.time()
threshold = 3 # this will be the z value
usage = ('\nusage: python search_bwt.py [--no-indels] [test|<reference file name>] [<read file name>]\n')
if '--no-indels' in sys.argv:
global NO_INDELS
NO_INDELS = True
if '--linear-gaps' in sys.argv:
global gap_open, gap_ext
gap_open = 0
gap_ext = 1
if len(sys.argv) == 1:
print usage
return
elif sys.argv[1].lower() == 'test':
test()
return
elif len(sys.argv) < 3:
print usage
return
for i in range(0,len(sys.argv)):
if sys.argv[i] == '-t' and i < len(sys.argv)-2:
threshold = int(sys.argv[i+1])
fread = open(sys.argv[-1])
fref = open(sys.argv[-2])
ref = ''.join(fref.readlines()).replace('\n','')
read = ''.join(fread.readlines()).replace('\n','')
# estimate the subsitution matrix
if not '--no-sub-mat' in sys.argv:
global sub_mat
sub_mat = estimate_substitution_mat(ref,read)
sa = suffix_array(ref)
bw = bwt(ref)
bwr = bwt(ref[::-1])
print read
print_output(inexact_search(bw,bwr,read,threshold), sa, ref)
elapsed = time.time() - start
if '--show-time' in sys.argv: print 'time elapsed: '+str(elapsed)
if '--count-prunes' in sys.argv: print str(num_prunes) + ' nodes pruned.'
print "error score upper bound: " + str(threshold)
fread.close()
fref.close()
def main():
start = time.time()
threshold = 3 # this will be the z value
usage = '\nusage:\tpython search_bwt.py [--no-indels] <reference file name> <read file name>' \
'\nor:\tpython search_bwt.py test\n'
if '--no-indels' in sys.argv:
global NO_INDELS
NO_INDELS = True
if '--linear-gaps' in sys.argv:
global gap_open, gap_ext
gap_open = 0
gap_ext = 1
if len(sys.argv) == 1:
print usage
return
elif sys.argv[1].lower() == 'test':
test()
return
elif len(sys.argv) < 3:
print usage
return
for i in range(0, len(sys.argv)):
if sys.argv[i] == '-t' and i < len(sys.argv) - 2:
threshold = int(sys.argv[i + 1])
fread = open(sys.argv[-1])
fref = open(sys.argv[-2])
ref = ''.join(fref.readlines()).replace('\n', '')
read = ''.join(fread.readlines()).replace('\n', '')
# estimate the substitution matrix
if '--no-sub-mat' not in sys.argv:
global sub_mat
sub_mat = estimate_substitution_mat(ref, read)
sa = suffix_array(ref)
bw = bwt(ref)
bwr = bwt(ref[::-1])
print read
print_output(inexact_search(bw, bwr, read, threshold), sa, ref)
elapsed = time.time() - start
if '--show-time' in sys.argv:
print 'time elapsed: ' + str(elapsed)
if '--count-prunes' in sys.argv:
print str(num_prunes) + ' nodes pruned.'
print "error score upper bound: " + str(threshold)
fread.close()
fref.close()
def test():
#s = 'ATGCGTAATGCCGTCGATCG'
s = 'CGATCCGCGCTGCTGATGATCGATG'
read = 'GATGAT'
threshold = 2
sa = suffix_array(s)
bw = bwt(s)
bwr = bwt(s[::-1])
print_output(inexact_search(bw,bwr,read,threshold), sa, s, read)
def test():
# Another test: 'ATGCGTAATGCCGTCGATCG'
s = 'CGATCCGCGCTGCTGATGATCGATG'
read = 'GATGAT'
threshold = 2
sa = suffix_array(s)
bw = bwt(s)
bwr = bwt(s[::-1])
print_output(inexact_search(bw, bwr, read, threshold), sa, s, read)
def compress_fasta_command(filename, outputfile):
"""
Ecrit la séquence compressée du fichier FASTA dans un fichier de sortie.
"""
seq = FastaFile(filename).complete_sequence()
bwt_seq = bwt(seq) # BWT(sequence)
mtf_seq = move_to_front(bwt_seq) # MTF(BTW(sequence))
rle_seq = rle(mtf_seq) # RLE(MTF(BTW(sequence)))
freq = occurences(rle_seq)
huffman_code = encode(freq) # création de la table de codage
output_string = ""
d = {c[0]:c[1] for c in huffman_code} # conversion de la table de codage en dictionnaire
for c in rle_seq:
output_string += d[c] # HUFFMAN(RLE(MTF(BTW(sequence))))
with open(outputfile, 'wb') as out: # ouverture en format binaire
for i in range(0, len(rle_seq), 8): # écriture par paquet d'octet
out.write(hex(int(output_string[i:i+8], 2)))
def test_bwt(sample_str):
assert bwt(sample_str + '\xff') == bwt_naive(sample_str + '\xff')
def test_bwt_simple():
assert bwt('abacaba\xff') == list('\xffcbbaaaa')
def test_ibwt(sample_str):
dollar_sign = '\xff'
bwt_of_str = bwt(sample_str + dollar_sign)
assert ibwt_string(bwt_of_str, dollar_sign) == sample_str + dollar_sign
